Dogs 'can differentiate between happy and angry faces'
Half the dogs were rewarded for touching an image of a happy face, while the other half were rewarded for touching a picture of an angry face.
According to new research, dogs can tell the difference between happy and angry human faces - a fact which may come as no surprise to dogs owners.
Scientists from the Messerli Research Institute's Clever Dog Lab in Vienna trained dogs to link pictures of happy and angry faces with a reward, BBC News has reported.
Twenty dogs were repeatedly shown half-pictures - either of the lower mouth or upper eye area of happy and angry human faces.
Half of the dogs were given a treat when their nose touched the happy pictures, while the other half received a treat when their nose touched the angry pictures.
Following this, researchers carried out a number of tests on just over half of the dogs (who were deemed to have learned the task well enough).
Lead researcher Ludwig Huber told BBC News that in one test condition, the dog were shown pictures of new faces that they had not seen in training. In another, they were shown different parts of the same faces.
Dr Kun Guo from the University of Lincoln told the BBC: "Showing dogs only half of the face and then the other half separately means they can't rely on the shape of the eyes or mouth - they must have some sort of template in their mind.
"So it looks like they can really discriminate between happy and angry."
However, the findings, which have been published in Current Biology, do not show that the dogs actually understand the meaning behind these expressions, he added.
While Dr Huber agreed with this, he noted that the group of dogs who had to touch the angry face during training took three times longer to learn the task than the other group.
"So here we have some suggestive evidence that they interpret those pictures, and maybe they really understand an angry face to be something they don't like," he told BBC News.
BSAVA is to partner with BVA Live (11-12 June 2026) to champion clinical research.
